Meyhem Researcher

Multi-query research tool with LLM-ranked results and freshness control. Breaks a topic into focused queries, previews top results. No API key.

Security Scan

The skill is internally consistent with its stated purpose (a Python tool that issues searches to api.rhdxm.com); its requirements and instructions match its behavior, but it does transmit user queries and selected URLs to a third-party service so avoid sending sensitive data.

Purpose & Capability

Name/description (multi-query research with LLM-ranked results) aligns with included code and SKILL.md: the Python script issues HTTP requests to api.rhdxm.com, breaks topics into sub-queries, and previews results. Required binary (python3) and lack of credentials are appropriate.

Instruction Scope

SKILL.md and researcher.py explicitly send the user's query, agent_id, and selected URLs to https://api.rhdxm.com (including a follow-up select call to fetch preview content). The README warns not to include sensitive or proprietary content; this network transmission is within the skill's stated purpose but is the primary privacy risk.

Install Mechanism

No install spec; the skill is instruction-only with a small Python script. No downloads or archive extraction—lowest install risk.

Credentials

The skill requests no environment variables, no credentials, and accesses only argv. This is proportionate to its functionality. Note: the agent_id argument can leak identifying information if set to an organization-specific value.

Persistence & Privilege

The skill does not request always:true and does not modify system or other skills' configuration. It runs only when invoked.

Assessment

This skill behaves as advertised: it sends your queries, an agent identifier, and any selected URLs to a third-party API (api.rhdxm.com). Before installing or using it: (1) Don't include secrets, proprietary text, or internal identifiers in queries or the agent_id. (2) Consider whether sending search terms and clicked URLs to an external service is acceptable for your data policy. (3) If you need stronger privacy, review the GitHub source yourself or run a local/offline research workflow instead. (4) Test with non-sensitive queries first and inspect returned previews/URLs for unexpected behavior.

Meyhem Deep Researcher

Multi-query research tool. Breaks a topic into focused queries, searches via api.rhdxm.com with LLM-ranked results, and previews top results.

No API key. No signup. No rate limits.

Why Meyhem Researcher?

• LLM-ranked results: every result scored for relevance by an LLM prior
• Freshness control: request real-time, hourly, daily, or weekly freshness
• Multi-query workflow: break a topic into multiple queries, search, preview top results

Quick Start

python3 researcher.py "transformer attention mechanism"
python3 researcher.py "kubernetes networking" -n 3 -q 5
python3 researcher.py "AI regulation 2026" --freshness realtime
python3 researcher.py "climate policy updates" --freshness hour --agent my-researcher

Quick Start (REST)

Full API docs: https://api.rhdxm.com/docs

curl -s -X POST https://api.rhdxm.com/search \
  -H 'Content-Type: application/json' \
  -d '{"query": "YOUR_QUERY", "agent_id": "my-researcher", "max_results": 10, "freshness": "hour"}'

Freshness options: realtime (never cached), hour, day, week, or omit for default (1hr cache).

MCP

You can also connect via MCP at https://api.rhdxm.com/mcp/ for richer integration.

Data Transparency

This skill sends your search query, an agent identifier, and any selected URLs to api.rhdxm.com. The skill does not access local files, environment variables, or credentials on its own, but anything you include in the query or agent_id will be transmitted. Avoid sending sensitive or proprietary content.

Source code: https://github.com/c5huracan/meyhem